Freshwater Biology (2011)
doi:10.1111/j.1365-2427.2010.02569.x
Spatiotemporal heterogeneity of soil and sediment
respiration in a river-floodplain mosaic
(Tagliamento, NE Italy)
MICHAEL DOERING*, URS UEHLINGER*, THEKLA ACKERMANN*, MICHAEL WOODTLI*
A N D K L E M E N T T O C K N E R * , †, ‡
*EAWAG, Swiss Federal Institute of Aquatic Science and Technology, Dübendorf, Switzerland
†
Institute of Integrative Biology, Swiss Federal Institute of Technology, ETH, Zurich, Switzerland
‡
IGB, Leibniz-Institute of Freshwater Ecology and Inland Fisheries, Berlin, and Institute of Biology,
Freie Universität Berlin, Germany
SUMMARY
1. In their natural state, river floodplains are composed of a complex mosaic of contrasting
aquatic and terrestrial habitats. These habitats are expected to differ widely in their
properties and corresponding ecological processes, although empirical data on their
capacity to produce, store and transform organic matter and nutrients are limited.
2. The objectives of this study were (i) to quantify the spatiotemporal variation of
respiration, a dominant carbon flux in ecosystems, in a complex river floodplain, (ii) to
identify the environmental drivers of respiration within and among floodplain habitat
types and (iii) to calculate whole-floodplain respiration and to put these values into a
global ecosystem context.
3. We measured soil and sediment respiration (sum of root and heterotrophic respiration;
SR) throughout an annual cycle in two aquatic (pond and channel) and four terrestrial
(gravel, large wood, vegetated island and riparian forest) floodplain habitat types in the
island-braided section of the near-natural Tagliamento River (NE Italy).
4. Floodplain habitat types differed greatly in substratum composition (soil to coarse
gravel), organic matter content (0.63 to 4.1% ash-free dry mass) and temperature (seasonal
range per habitat type: 8.6 to 33.1 C). Average annual SR ranged from 0.54 ± 1.56
(exposed gravel) to 3.94 ± 3.72 lmol CO2 m)2 s)1 (vegetated islands) indicating distinct
variation in respiration within and among habitat types. Temperature was the most
important predictor of SR. However, the Q10 value ranged from 1.62 (channel habitat) to
4.57 (riparian forest), demonstrating major differences in habitat-specific temperature
sensitivity in SR.
5. Total annual SR in individual floodplain habitats ranged from 160 (ponds) to
1205 g C m)2 (vegetated islands) and spanned almost the entire range of global ecosystem
respiration, from deserts to tropical forests.
Keywords: aquatic–terrestrial metabolism, ecosystem process, environmental heterogeneity, floodplain
Introduction
Correspondence: Michael Doering, EAWAG, Swiss Federal
Institute of Aquatic Science and Technology, 8600 Dübendorf,
Switzerland. E-mail: [email protected]
2011 Blackwell Publishing Ltd
River floodplains, i.e. the active tract and fringing
riparian zone along rivers, can be portrayed as a
complex mosaic of contrasting aquatic and terrestrial
1
2
M. Doering et al.
a
gn
e
Mt
go
Ra
an
ari
Rip rest
fo
tive
in
Ac
pla
od
Flo
ct
tra
habitats. This spatially and temporally complex habitat
mosaic varies widely in habitat properties, such as
habitat composition (e.g. different soil and sediment
types), temperature patterns, hydrological conditions
(e.g. water content) and in terms of ecosystem processes (e.g. respiration). Thus, floodplains may serve as
a model to study the effect of habitat properties on
ecosystem processes in aquatic, semi-aquatic and
terrestrial habitats (Naiman, Décamps & McClain,
2005; Ward et al., 2002; Tockner, Lorang & Stanford,
2010; Fig. 1). For example, exposed gravel is considered as a harsh environment exhibiting extreme temperature variation, high water stress and low
productivity (Tockner et al., 2006), whereas riparian
forests provide benign environmental conditions and
are rich in resources that sustain high productivity
(Tockner & Stanford, 2002). Vegetated islands and
riparian forests provide large amounts of bioavailable
organic carbon to aquatic habitats, where it is either
processed in situ or exported downstream (Cuffney,
1988; Junk, Bayley & Sparks, 1989; Langhans et al.,
2006, 2008). However, most ecosystem studies have
focussed on either aquatic (Pusch & Schwoerbel, 1994;
Naegeli & Uehlinger, 1997; del Giorgio & Williams,
2005; Ingendahl et al., 2009) or terrestrial (Davidson,
Belk & Boone, 1998; Buchmann, 2000; Euskirchen et al.,
2003; Ball et al., 2009) habitat types. A key challenge in
ecosystem research is to understand the influence of
habitat properties on processes. Therefore, we need to
study the combined effects of the composition, spatial
configuration and dynamics of both aquatic and
terrestrial habitat types on ecosystem processes.
In this study, we focussed on the effect of habitat
properties on sediment and soil respiration in aquatic
and terrestrial river-floodplain habitats. Respiration is
a central ecosystem process that regulates organic
matter decomposition, detrital storage, nutrient
cycling and flux of CO2 to the atmosphere (Andrews
& Schlesinger, 2001). Therefore, respiration can be
used as an indicator of ecosystem integrity (Van Voris
et al., 1980; Uehlinger & Naegli, 1998; Euskirchen
et al., 2003). Generally, respiration increases exponentially with temperature and is often limited by deficits
in soil moisture and substratum availability (Buchmann, 2000; Euskirchen et al., 2003; Tang et al., 2006).
Thus, even marginal changes in temperature, water
availability or quantity and quality of organic matter
probably alter aquatic and turnover rates of terrestrial
carbon (Schimel et al., 1994; Jones, 1995; Randerson
Flow direction
0 250
Cropland
Riparian forest
Exposed gravel sediments
Islands
500
750
m
1000
Pond
Channel
Fig. 1 Aerial photograph (May 2005) of the main study area
(upper panel) and detailed map of the study area showing the
spatial distribution of individual habitat types (lower panel).
Large wood accumulations are too small to be shown on the
map. All habitats were delineated by an intensive field survey in
February 2005 at low discharge using a dGPS.
et al., 1996; Davidson et al., 2000; Xu & Qi, 2001; Li, Xu
& Zou, 2006; McIntyre et al., 2009).
To gain a better understanding of the factors
governing soil respiration in a spatially complex and
temporally dynamic ecosystem, we conducted intensive field studies in the island-braided section of one
of the last remaining near-natural river corridor in
Central Europe (Tagliamento River, NE Italy). The
Tagliamento offers insights into the structural and
functional complexity that probably characterised
most Alpine rivers before river engineering and
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
Sediment respiration in a floodplain mosaic
hydrological power production affected river morphology, sediment balance and hydrology. We measured soil temperature, soil water content, grain size
distribution, organic matter content and soil and
sediment CO2 efflux in two aquatic and four terrestrial habitat types over an annual cycle. The main
goals of the study were to (i) quantify the spatiotemporal variability of soil and sediment respiration (i.e.
CO2 flux as the sum of root and heterotrophic
respiration; SR) of the dominant aquatic and terrestrial habitat types, (ii) identify the main potential
environmental drivers of SR and (iii) quantify total
annual SR at the entire floodplain scale and to
compare SR with respiration reported from various
ecosystem types globally.
Methods
Study site
This study was conducted from January to October
2005 in the island-braided section of the seventh-order
Tagliamento River in north-eastern Italy (46 N, 1230¢
E). The Tagliamento (catchment area: 2580 km2) originates at 1000 m a.s.l. in the southern fringe of the
European Alps and flows almost unimpeded by dams
for 172 km to the Adriatic Sea.
The main study area was a 116-ha island-braided
floodplain complex (river km 79.8–80.8; 135 m a.s.l.)
where maximum annual water level fluctuations were
about 2 m (Tockner et al., 2003). This reach contains a
spatially complex and temporally dynamic habitat
mosaic dominated by extensive areas of exposed
riverine sediments (Petts et al., 2000; Van der Nat
et al., 2003) (Fig. 1). The 800-m-wide active tract of the
floodplain encloses numerous islands. The northwestern border of the active tract is fringed by a
ribbon of intact riparian forest, with black poplar
(Populus nigra L.) and five willow (Salix spp.) species
as the dominant tree species (Karrenberg et al., 2003).
The steep hillslope forest of Monte Ragogna fringes
the south-eastern border of the floodplain (Fig. 1). Soil
types of the riparian forest and vegetated islands are
classified as Eutrochrept fluvisols.
At base flow (about 20 m3 s)1), the floodplain is
composed of a complex mosaic of exposed gravel
habitats (60.3 ha, 51.8% of total area), vegetated
islands (10.4 ha, 8.9%), large wood accumulations
(0.4 ha, 0.4%), the channel network (18.2 ha, 15.6%),
3
numerous ponds (0.6 ha, 0.6%) and the riparian forest
(26.5 ha, 22.8%) (Table 1). The river has a flashy flow
regime (Q80 = 72 m3 s)1; Ward et al., 1999) with frequent and short flow and flood pulses (Tockner,
Malard & Ward, 2000) throughout the year (Arscott
et al., 2002). Detailed information on the Tagliamento
catchment and the main study area has been published
by Ward et al. (1999), Gurnell et al. (2001), Tockner et al.
(2003), Bertoldi et al. (2009) and Tonolla et al. (2010).
Characterisation of habitat patches
Two aquatic (channels and ponds) and four predominantly terrestrial (riparian forest, vegetated islands,
large wood accumulations and exposed gravel) habitat types (Fig. 1; Table 1) were analysed for grain size
distribution, water content and organic matter content
(four times over the annual cycle). The number of
sampling sites was allocated according to the relative
areal contribution of each individual habitat type to
Table 1 Characterisation of the terrestrial and aquatic habitat
types sampled in this study. LWA = Large wood accumulation.
Apart from riparian forests, all habitats are located within the
active tract of the floodplain
Habitat type
Characteristics
Channels
Permanent lotic primary and secondary
channels composed of coarse permeable
gravel sediments and typically fringed
by gravel bars
Permanent parafluvial ponds, often located
adjacent to islands and large wood
accumulations, composed of coarse
permeable gravel sediments
Predominantly forested terrestrial* habitat
type, fringing the active tract of the
floodplain and characterised by developed
soil (Eutrochrept fluvisol). The vegetation
is mainly composed of black poplar
(Populus nigra) and different willow
(Salix spp.) species
Predominantly terrestrial* habitat type,
mainly colonised by Populus nigra and
different willow species and characterised
by sandy substrata and developed
soil (Eutrochrept fluvisol).
Predominantly terrestrial* accumulations
of large wood trapping mainly fine
sediments and organic matter
Predominantly terrestrial* areas
characterised by bare or sparsely
vegetated gravel deposits
Ponds
Riparian forest
Islands
LWA
Gravel
*These habitats become aquatic habitats during floods only.
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
4
M. Doering et al.
the whole-floodplain area. Samples were taken randomly but evenly distributed along the riparian forest
strip on the north-western part of the floodplain,
along the main and site channel and in the exposed
gravel. Individual vegetated islands, large wood
accumulations and ponds were subject to sampling
within the whole active tract. The water content in
terrestrial soils and sediments was always below
maximum water-holding capacity. The aquatic sites
were characterised by coarse and permeable gravel
deposits (Table 2). Therefore, anaerobic conditions are
expected to be absent in this floodplain system. Soil
(riparian forest, vegetated islands) and sediment
(large wood accumulations, channels, ponds and
exposed gravel) water content (% of dry soil and
sediment) was determined gravimetrically 3–12 h
after sample collection. Samples were weighed, dried
at 105 C for 24 h and reweighed. Gravimetric water
content (WC; % of dry sediment) of aquatic sediments
was calculated according to eqn 1.
WC ¼
P Vtot q
100
S þ P Vtot q
ð1Þ
where P is porosity, Vtot is sediment volume in cm3, S
is dry sediment weight, and q is water density.
Porosity was assumed to be 0.2 (Eglin, 1990; Jussel,
1992) and water density 1 g cm)3 (Eglin, 1990).
To measure the maximum water-holding capacity
(WHC; % of dry soil and sediment), five random
samples per habitat type were dried at 105 C,
weighed, saturated with deionised water for 24 h,
drained for 24 h and weighed again. Organic matter
content of dry samples was measured as ash-free dry
mass (AFDM; ashed at 500 C for 3 h) per kg dry soil
or sediment. Ashed sediments were sieved to separate
grain size fractions <0.063 mm, 0.063–2 mm, 2–4 mm,
4–8 mm and >8 mm.
Temperature was continuously recorded in terrestrial and aquatic habitat types. Five temperature
loggers (DS1921G, Dallas Semiconductor, Dallas,
U.S.A.) were deployed in each terrestrial habitat type
at a depth of 5 cm. Five loggers (TR MINILOG,
VEMCO, Nova Scotia, Canada and PDLR70; Ecotech,
Germany) were also deployed at the sediment surface
of each aquatic habitat.
Measurement of SR in terrestrial habitats
Soil and sediment respiration (i.e. sum of root and
heterotrophic respiration; SR) was measured as
positive CO2 flux per m)2 s)1 in January, April, July
and October 2005 using a soil respiration chamber (Li
6400, LiCor, Lincoln, Nebraska, U.S.A.) attached to a
portable Li-6400 infrared gas analyser (IRGA). PVC
collars of known volume (8 cm long, 10.5 cm inside
diameter) were inserted into the soil evenly distributed along the north-western contiguous riparian
forest stretch (n = 16 collars per date), in the soils of
Table 2 Characterisation of terrestrial and aquatic habitat types. Average and standard deviation of pooled data from four seasons.
Temperature includes spot measurements (C), water content is expressed as % of dry sediments, water content as % of the waterholding capacity, organic matter (OM) content of two size fractions and total organic matter content per mass of sediment as g AFDM
kg)1. Grain size classes are expressed in % of the sum of all size fractions
Habitat
Temperature (C)
Water content (%)
% of water-holding capacity
OM >2 mm (g AFDM kg)1)
OM <2 mm (g AFDM kg)1)
Total OM (g AFDM kg)1)
Grain size distribution
>8 mm (%)
8–4 mm (%)
4–2 mm (%)
2–0.063 mm (%)
<0.063 mm (%)
Riparian forest
n = 63
Island
n = 114
LWA
n = 52
12.9
25.7
57.7
8.3
32.6
40.9
±
±
±
±
±
±
7.4
10.3
22.7
6.0
12.4
15.64
14.2
15.2
36.6
5.4
17.9
23.3
±
±
±
±
±
±
8.9
10.5
24.2
5.7
12.4
16.4
14.9
15.7
55.8
3.7
15.5
19.2
±
±
±
±
±
±
0.04
0.06
0.8
80.1
19.0
±
±
±
±
±
0.07
1.4
1.3
17.2
17.5
0.06
0.04
0.5
87.4
12.0
±
±
±
±
±
0.2
0.1
0.7
15.3
15.4
3.4
1.8
1.6
86.2
7.0
±
±
±
±
±
Gravel
n = 322
Channel
n = 100
Pond
n = 53
9.8
10.8
29.8
4.9
13.0
17.2
16.8
3.7
41.1
3.3
3.0
6.5
±
±
±
±
±
±
11.7
8.7
31.3
2.2
2.7
2.9
12.9
9.4
100
3.6
2.8
6.3
4.6
0.8*
0
1.3
1.3
1.3
13.5
9.8
100
2.3
3.7
6.6
±
±
±
±
±
±
6.2
1.4*
0
1.6
2.3
1.9
9.2
3.2
2.0
13.2
5.2
46.1
8.6
5.1
39.2
1.0
±
±
±
±
±
26.7
5.6
4.0
31.9
1.9
46.1†
20.1 ±
11.7 ±
21.8 ±
0.3 ±
7.0
3.3
8.4
0.3
46.1†
16.0 ±
8.1 ±
29.4 ±
0.5 ±
8.6
4.4
12.2
0.4
±
±
±
±
±
±
AFDM, ash-free dry mass; LWA, large wood accumulation; n, number of samples in each habitat.
*Values calculated for permanent aquatic sediments.
†Estimate based on the grain size distribution of gravel habitats.
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
Sediment respiration in a floodplain mosaic
eight distinct vegetated islands (n = 29), in the
trapped sediments of 13 individual large wood
accumulations (n = 13) and in the sediments of the
exposed gravel sediment matrix (n = 81). To avoid
effects of prior sampling disturbance, sites were
located slightly offset from previous sampling sites
but within 1 m; the exact position of each randomly
chosen sampling location was determined using a
differential global positioning system (dGPS; accuracy
<0.5 m). Collars were inserted to a sediment ⁄ soil
depth of 5–7 cm. According to Norman et al. (1997),
the insertion of a collar may result in initially high
CO2 fluxes that stabilise after 10–30 min. To minimise
this effect, collars were installed at least 24 h prior to
measurement (Buchmann, 2000). The soil chamber
was set on top of the collar to measure the undisturbed CO2 flux. The IRGA measurement was
repeated four times per date and collar, and values
were averaged. To estimate diel variation in CO2
fluxes, five additional collars were exposed at each
habitat type and measured three to four times from 7
am to 8 pm. Temperature was measured at 5-cm
sediment depth next to the collar. After each measurement, the soil enclosed by each collar was
collected, stored in sealed PVC bags and transported
to the laboratory for water, organic matter content and
grain size analysis.
Measurement of SR in aquatic habitats
Respiration of aerobic aquatic sediments was measured as the change in O2 concentration over time in
plexiglas tubes (5.2 cm diameter, 32 cm long) sealed
with rubber stoppers (Uehlinger, Naegeli & Fisher,
2002). Surface autotrophs (exclusively epilithic algae)
were excluded. This measure corresponds to the
respiration measurements in terrestrial habitats that
did not include the respiration of trees or shrubs.
Sediment samples evenly distributed in the main and
side channel (n = 25 sampling sites per date) and in 14
individual ponds (n = 14) were collected from 0 to
20 cm depth. Prior to sampling, the surface layer,
typically covered with epilithic algae, was removed.
Respiration tubes were filled with pre-sieved sediments (£8 mm), filled to the top with surface water
from the sampling site and subsequently incubated in
situ, i. e. tubes were buried in surface sediments at the
sampling site for about 4 h (Uehlinger et al., 2002).
Sediment particles >8 mm were excluded because the
5
random presence or absence of a few pebbles, representing a relatively large inactive metabolic volume in
the relatively small chambers, result in high variability of oxygen consumption within habitats. Oxygen
concentration and temperature were measured with a
portable oxygen metre (Oxi 340 ⁄ bset, WTW, Weilheim, Germany) at the beginning and end of the
incubation period. After incubation, sediments were
stored frozen and later analysed for organic matter
content and grain size distribution.
Calculation of SR from sediment tube measurements
Based on the oxygen consumption in the chamber
water r (g O2 m)3 h)1), we first calculated the respiration per sediment volume Rv (lmol CO2 m)3 s)1)
according to eqn 2:
Rv ¼ r 106
1
Vw
RQ
32 3600 Vc
ð2Þ
Vw is the volume of water in a chamber (m3), Vc the
volume of sediment (including porosity) filled into a
chamber (m3), and RQ is respiratory quotient (0.85;
Dilly, 2001). The porosity of exposed gravel, channel
and pond sediments was assumed to be 20% (Eglin,
1990; Jussel, 1992). To get an estimate of areal
respiration R (lmol CO2 m)2 s)1) for the uppermost
cm of pond or channel sediments, we multiplied Rv by
0.2 m and by 0.54. The factor 0.54 considers that size
fraction >8 mm accounted on average for 54% of the
sediment volume (this percentage was determined by
analysing 322 gravel samples collected at 81 sites).
Temperature dependence of SR
We described the temperature dependence of respiration as (3):
y ¼ a expðbTÞ ;
ð3Þ
where T is the sediment temperature (C) at 5 cm
depth (terrestrial habitats) or the temperature of the
chamber water. The model parameters a and b were
identified using nonlinear regression applied to respiration and temperature data of each individual
habitat type over the course of the study.
Residual analyses were performed to test for correlations between residuals of the temperature model
and soil and sediment water content, organic matter
content and individual grain size fractions.
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
6
M. Doering et al.
Q10 values were calculated according to eqn 4
(Buchmann, 2000):
Q10 ¼ exp10b
ð4Þ
Respiration at reference temperature (Tr) 12 C was
calculated according eqn 5:
RðTr Þ ¼ expbðTTr Þ
ð5Þ
Annual soil and sediment CO2 flux
The total annual SR for the individual habitat types, as
well as for the entire floodplain system, was calculated based on the exponential equations derived
from empirical field data (Table 5). Temperature data
were taken from average daily logger data. The areal
extent of each habitat type was used to scale up the
results to the entire floodplain system and was
extracted from a geographic information system
(GIS) layer based on a dGPS mapping survey in
2005 (Fig. 1).
Statistical analyses
A factorial design was used with habitat type (six
levels) and season (four levels) as factors and respiration as the dependent variable. The number of
replicates of each habitat type was partly determined
by its coverage and distribution within the floodplain
section (i.e. by a single contiguous forest stretch or
gravel matrix, and a limited number of islands and
channels). Since subsequent seasonal measurements
were placed next to the prior sampling sites, two-way
A N O V A with a posteriori tests (Bonferoni post hoc) was
applied to separate means. Data were log-transformed, if necessary, to meet the assumptions of
normality (Sokal & Rohlf, 1995). Spearman rank
correlation was applied to describe the relationships
between habitat variables and within-habitat respiration. Regression analysis was carried out to predict
soil respiration from diel and seasonal temperature
variation. All statistics were calculated using Statistica
6.0 (Statsoft Inc., Tulsa, Oklahoma, U.S.A.).
Results
Environmental characterisation of habitat types
Each habitat type exhibited distinct seasonal and diel
temperature patterns (Fig. 2). The seasonal varia-
tion in daily mean temperature ranged from 8.6 C
(channel) to 33.1 C (exposed gravel). The diel amplitude, averaged over the annual cycle, ranged between
2.7 C (channel) and 9.7 C (exposed gravel) (Fig. 2).
The total annual degree days (sum of average daily
temperature) ranged from 4017 (riparian forest) to
4560 (exposed gravel).
Particles >8 mm prevailed in exposed gravel sediments and in aquatic habitats (>46%), whereas the
grain size fraction 2–0.063 mm dominated other types
of terrestrial habitats (>39.2%). Average gravimetric
water content ranged from 3.7% (exposed gravel
sediments) to 25.7% (riparian forest). The average
WC, expressed as % of WHC, was lowest in vegetated
islands (36.6%) and highest in the riparian forest
(57.7%). In the terrestrial habitats, the total organic
matter per mass of sediment (TOM; g AFDM kg)1)
ranged from 6.5 g (0.65%) in exposed gravel sediments to 40.9 g (4.1%) in the riparian forest, with the
grain size fraction <2 mm as the dominant organic
matter (OM) fraction. The OM content in aquatic
habitats was similar to the content in exposed gravel
sediments (Table 2).
Spatiotemporal variation of SR
Average annual SR (lmol CO2 m)2 s)1) per habitat
type ranged from 0.54 ± 1.56 lmol (exposed gravel
sediment) to 3.94 ± 3.72 lmol (vegetated island). In
particular, exposed gravel sediments revealed a
patchy spatial distribution in SR indicated by high
standard deviations. Seasonally, average SR varied
from 0.04 (January) to 1.29 lmol CO2 m)2 s)1 (July) in
exposed gravel sediments and from 0.86 (January) to
8.22 lmol CO2 m)2 s)1 (July) in vegetated islands.
Seasonal variation in SR was relatively low in aquatic
habitats (Table 3). SR differed significantly among
habitat types (d.f. = 5; F = 169.58; P < 0.001), although
SR in the riparian forest and vegetated islands was not
significantly different (Bonferoni post hoc; P > 0.05).
SR varied significantly with time (season) (d.f. = 3;
F = 121.06; P < 0.001) and there was a significant
interaction between time and habitat type (d.f. = 15;
F = 10.31; P < 0.001).
Temperature-corrected average annual SR (reference
temperature: 12 C, eqn 5, lmol CO2 m)2 s)1) ranged
from
0.21 ± 0.50 lmol
(exposed
gravel)
to
2.66 ± 2.04 lmol (riparian forest). Average seasonal
SR varied in exposed gravel sediments from 0.07 lmol
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
Sediment respiration in a floodplain mosaic
50
Avr 12.2 ± 8.0
Min –4.0
Max 27.3
Gravel
Avr 12.5 ± 9.4
Min –2.8
Max 30.3
Riparian forest
Avr 11.0 ± 6.9
Min 0.2
Max 22.4
Channel
Avr 12.4 ± 2.4
Min 7.1
Max 15.7
Island
Avr 11.2 ± 8.2
Min –1.8
Max 24.9
Pond
Avr 13.7 ± 5.1
Min –3.4
Max 22.6
40
30
7
20
10
0
–10
50
Air temperature
40
30
Temperature (°C)
20
10
0
–10
50
40
30
20
10
No Data
0
–10
50
Avr 12.2 ± 8.7
Min –2.4
Max 27.6
Large wood accumulation
40
30
Feb
Apr
Jun
Aug
Oct
Dec
Feb
20
10
0
–10
Feb
Apr
Jun
Aug
Oct
Dec
Feb
Fig. 2 Average daily air temperature measured 10 km south-east from the study site (Fagagna; 461004¢N, 130903¢E) and average,
minimum and maximum daily temperature retrieved from loggers including summarising statistics (average ± SD, minimum and
maximum of annual daily mean temperature in C) in the six aquatic and terrestrial habitat types from January 2005 to February 2006
(arrows indicate seasonal sampling dates; see text for further information on sampling sites).
Table 3 Seasonal soil and sediment respiration (lmol CO2 m)2 s)1) in aquatic and terrestrial habitats (average X ± SD)
Riparian forest
n = 63
Islands
n = 114
Month
X ± SD
CV
X ± SD
CV
X ± SD
CV
X ± SD
CV
X ± SD
CV
X ± SD
CV
January
April
July
October
0.97
3.10
7.44
2.67
53
31
27
51
0.86
3.13
8.22
3.38
133
43
46
94
0.37
1.50
5.51
1.46
146
57
42
86
0.04
0.52
1.29
0.30
573
158
206
194
0.57
1.00
1.03
0.62
56
59
52
81
0.23
0.47
1.09
0.62
83
60
42
60
±
±
±
±
0.51
0.96
2.04
1.36
±
±
±
±
1.14
1.35
3.79
3.17
LWA
n = 52
±
±
±
±
Gravel
n = 322
0.54
0.85
2.34
1.26
±
±
±
±
0.24
0.82
2.65
0.58
Channels
n = 100
±
±
±
±
0.32
0.59
0.54
0.5
Pond
n = 53
±
±
±
±
0.19
0.28
0.46
0.37
LWA, large wood accumulation; CV, coefficient of variation.
(January) to 0.38 lmol (July) and in the riparian forest
from 1.64 lmol (April) to 5.01 lmol (January). In
terrestrial habitats, temperature-corrected SR peaked
in July, except in the riparian forest, and was most
variable in January. In aquatic habitats, temperaturecorrected SR remained fairly constant throughout the
year (Fig. 3). Temperature-corrected average annual
SR (lmol CO2 m)2 s)1) per kg OM ranged from
408.5 ± 835.4 (Islands) to 65.3 ± 221.3 (Gravel). The
differences between habitats were significant (d.f. = 5;
F = 171.51; P < 0.001) but an a posteriori test indicated
significant differences between gravel and all other
habitats only (Bonferroni post hoc; P < 0.001). Seasonal
differences in temperature-corrected SR per kg OM
were significant (d.f. = 3; F = 7.88; P < 0.001) but
mainly because of high values in riparian forest, islands
and LWA habitats in January (Bonferroni post hoc;
P < 0.001). The calculated Q10 values ranged from 1.62
(channel habitats) to 4.57 (riparian forest) (Table 5).
Within habitat types, average spatial variation in
SR, expressed as the coefficient of variation (CV), was
lowest in the riparian forest (CV = 41%) and highest
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
8
M. Doering et al.
OM content) and SR, we found the strongest correlations between temperature and SR (0.32 < r < 0.85),
except for the channel habitat type (Table 4). The
relationship between temperature and SR was best
described by an exponential equation. Temperature
explained between 38 and 72% of the SR variation,
except for exposed gravel and channel habitats
(R2 = 0.05 and 0.11, respectively); although all models
were highly significant (P < 0.0001) (Table 5). Residual analysis confirmed temperature as the dominant
variable explaining respiration within individual
habitat types.
Respiration (µmol CO2 m–2 s–1)
8
Riparian forest
Island
Large wood accumulation
Gravel
Channel
Pond
7
6
5
4
3
2
1
0
January
April
July
October
Fig. 3 Total soil and sediment respiration (average ± SD) in
individual habitat types standardised at 12 C. N = 52–322 per
habitat type (see Table 3).
Annual soil and sediment CO2 flux
The habitat-specific annual SR (February 2005 to
January 2006) ranged from 160 (ponds and exposed
gravel sediments) to 1205 g C m)2 year)1 (islands),
thereby spanning almost the entire range of respiration reported from various ecosystem types worldwide (Tables 6 & 7). Total annual SR (area-weighted
mean) for the entire 116-ha island-braided floodplain
was calculated as 467 g C m)2 year)1, corresponding
to a total carbon efflux of 544 tC. The riparian forest
(23% of floodplain area) and island (9% of floodplain
area) contributed disproportionately to total annual
floodplain respiration (47 and 23%, respectively),
while the relative contribution of exposed gravel
sediments was low (18%) compared to its areal extent
(51% to total area; Table 6).
in exposed gravel sediments (282%). The withinhabitat spatial variation decreased with increasing
temperature and was therefore lowest in July
(Table 3). The diel variation in SR was highest in July
and ranged between 0.99 (vegetated island) and
1.63 lmol CO2 m)2 s)1 (large wood accumulation).
The diel amplitude in SR was significantly related to
the diel variation in temperature (r2 ‡ 0.82; P < 0.001).
Potential drivers of SR
Although significant correlations were observed for
several within-habitat variables (e.g. WC, WHC and
Table 4 Spearman rank correlations of total soil and sediment respiration (lmol CO2 m2 s)1) and different habitat variables (pooled
data from four seasons). Significant correlations (P £ 0.05) are highlighted in bold
Riparian
forest
n = 63
Islands
n = 114
Independent variable
r
P
r
P
r
P
r
Temperature (C)
Water content (%)
% of water-holding capacity
OM >2 mm (g kg)1sediment)
OM <2 mm (g kg)1sediment)
Total OM (g kg)1sediment)
Grain size >8 mm (%)
Grain size 8–4 mm (%)
Grain size 4–2 mm (%)
Grain size 2–0.063 mm (%)
Grain size <0.063 mm (%)
0.85
)0.37
)0.37
0.27
0.27
0.24
)0.43
)0.58
)0.22
)0.67
0.69
<0.001
0.002
0.002
0.036
0.030
0.054
<0.001
<0.001
0.077
<0.001
<0.001
0.72
)0.22
)0.22
0.10
0.10
0.10
)0.33
)0.42
0.02
)0.32
0.33
<0.001
0.018
0.018
0.273
0.286
0.301
<0.001
<0.001
0.866
<0.001
<0.001
0.73
)0.28
)0.28
0.12
0.18
0.17
0.19
0.18
0.15
)0.32
0.44
<0.001
0.046
0.046
0.194
0.391
0.224
0.169
0.190
0.297
0.020
0.001
0.35
0.17
0.17
0.22
0.08
0.21
)0.19
)0.19
)0.19
0.18
0.23
LWA
n = 52
Gravel
n = 322
Channels
n = 100
Ponds
n = 53
P
r
P
r
P
<0.001
0.002
0.002
0.148
<0.001
<0.001
0.001
0.001
0.001
0.001
<0.001
0.32
)0.19
–
0.27
0.13
)0.02
–
)0.03
)0.27
0.15
0.46
<0.001
0.099
–
0.001
0.210
0.830
–
0.742
0.006
0.132
<0.001
0.73
)0.18
–
0.14
0.19
)0.08
–
)0.00
0.01
)0.02
0.32
<0.001
0.278
–
0.394
0.175
0.555
–
0.990
0.945
0.868
0.018
OM, organic matter; LWA, large wood accumulation.
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
Sediment respiration in a floodplain mosaic
Table 5 Regression models between total
respiration (lmol CO2 m)2 s)1) and sediment (s; 5 cm depth) and water (w) temperatures calculated for six habitat types.
LWA = Large wood accumulation.
SE = Standard error. Q10 = Temperature
sensitivity. n = number of replicates
Table 6 Extend and respiration activity of
individual habitat types. Area (m2), area
(%), respiration (tC year)1), respiration
(g C m)2 year)1), respiration (%).
LWA = Large wood accumulation
9
R2
P
63
0.72
<0.0001
1.97
114
0.38
<0.0001
2.53
52
0.67
<0.0001
1.75
322
0.05
<0.0001
1.62
100
0.11
<0.0001
2.34
53
0.59
<0.0001
Habitat
Model
Parameter
SE
Q10
Riparian forest
R = a*exp(b*Ts)
R = a*exp(b*Ts)
LWA
R = a*exp(b*Ts)
Gravel
R = a*exp(b*Ts)
Channel
R = a*exp(b*Tw)
Pond
R = a*exp(b*Tw)
0.094
0.016
0.287
0.010
0.116
0.011
0.086
0.018
0.089
0.014
0.034
0.010
4.57
Island
a
b
a
b
a
b
a
b
a
b
a
b
Habitat
Area (ha)
Area (%)
Respiration
(tC year)1)
Respiration
(g C m)2 year)1)
Respiration
(%)
Riparian forest
Island
LWA
Channel
Pond
Gravel
Total floodplain
26.5
10.4
0.4
18.2
0.6
60.3
116.4
22.8
8.9
0.4
15.6
0.6
51.8
100.0
260
126
3
53
1
102
544
982
1206
624
292
160
168
467
47.8
23.1
0.5
9.7
0.2
18.7
100.0
Discussion
Soil and sediment respiration (SR) represents an
important carbon flux and thus is a well-studied
process in both permanent aquatic and terrestrial
ecosystems (del Giorgio & Williams, 2005; Luo & Zhou,
2006). However, very limited information is available
from morphologically complex and temporally
dynamic ecosystems such as river floodplains (but see
Valett et al., 2005). The floodplain of the Tagliamento
River, a model ecosystem of European importance,
provides an excellent framework to study the effect of
habitat properties on functional processes across
different habitat types over various temporal scales.
Our study clearly demonstrated that SR was highly
variable in space and time, reflecting the seasonal
dynamics and the spatial heterogeneity of the complex
floodplain mosaic. We identified temperature as the
main factor controlling SR within the various habitat
types, although the individual habitat types differed
considerably in their sensitivity to temperature.
Spatiotemporal variation of SR
Average SR differed among habitat types by an
order-of-magnitude (Fig. 3) and high standard devi-
=
=
=
=
=
=
=
=
=
=
=
=
0.305
0.152
1.292
0.068
0.394
0.093
0.165
0.056
0.423
0.048
0.168
0.085
n
ations revealed a patchy distribution of respiration
activity, especially for the exposed gravel sediments
(Table 3). The riparian forest and, in particular,
vegetated islands exhibited a high SR, as high as SR
reported for tropical forests (Table 7). The availability
of large pools of organic matter and the contribution
of root respiration to total SR probably explained
these high values. Microbial activity depends on the
availability of labile fractions of OM derived from
fresh litter input and in situ primary production
(Schimel et al., 1994; Janssens et al., 2001). In the
Tagliamento, islands are young (maximum age
20 years) and highly productive floodplain habitats
(Langhans et al., 2006, 2008). High input and storage
rates of bioavailable OM are expected to favour
heterotrophic SR on vegetated islands. Our results
underpin the potential role of vegetated islands as
‘hot spots’ for ecosystem processes (this study;
Langhans et al., 2006, 2008) as well as for biodiversity
(Gurnell et al., 2001; Tockner et al., 2006). Islands
exhibit a high perimeter to area ratio and are
expected to be linked with adjacent habitats that are
less productive and diverse. At the same time, islands
are among the first habitats that disappear as a
consequence of flow and channel regulation (Gurnell
et al., 2001). Therefore, they can be used as sensitive
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
10
M. Doering et al.
Table 7 Annual respiration (average ± SD; g C m)2 year)1) for
global terrestrial and aquatic ecosystem types (Bond-Lamberty
and Thomson, 2010; Raich and Schlesinger, 1992; Uehlinger,
1993), compared to aquatic and terrestrial annual respiration in
different floodplain habitat types of this study (in bold; *:
number of measurements as shown in Table 3)
Global soil ecosystems
and floodplain habitat types
Respiration
(g C m)2 year)1)
n
Arctic desert
Tundra
Northern bogs and mires
Arctic wetland
Arctic shrubland
Boreal grassland
Ponds
Gravel
Temperate savanna
Temperate desert
Boreal wetland
Boreal agriculture
Channels
Mediterranean savanna
Temperate shrubland
Boreal forest
Temperate wetland
Tropical wetland
Subtropical agriculture
Large wood accumulations
Temperate agriculture
Mediterranean agriculture
Subtropical grassland
Mediterranean grassland
Temperate streams
Temperate forest
Subtropical forest
Desert streams
Temperate grassland
Mediterranean forest
Riparian forest
Islands
Tropical forest
Tropical grassland
Tropical savanna
Tropical agriculture
1±
60 ±
94 ±
105 ±
116 ±
125 ±
160
168
181 ±
243 ±
271 ±
280 ±
292
439 ±
448 ±
448 ±
479 ±
500 ±
505 ±
624
650 ±
702 ±
721 ±
746 ±
769 ±
781 ±
806 ±
807 ±
827 ±
951 ±
982
1205
1284 ±
1326 ±
1437 ±
1525 ±
1
11
12
1
15
2
*
*
1
21
64
4
*
3
14
137
34
2
12
*
159
1
3
20
13
1020
35
9
215
41
*
*
227
23
13
23
0
20
55
0
113
94
0
263
159
243
85
289
218
390
262
186
362
0
311
480
316
429
434
668
512
413
605
895
520
657
indicators for assessing the integrity of floodplain
ecosystems.
In this study, we focussed solely on sediment
respiration – definitively an important ecosystem
process – and how this process is related to habitatspecific properties. However, these results cannot be
simply translated to other ecosystem processes. Langhans et al. (2008), for example, studied the decomposition of leaves, another important process in the
carbon cycle, in the same habitat types along the
Tagliamento. In contrast to SR, decomposition was an
order-of-magnitude higher in aquatic than in terrestrial habitats. Hence, a key challenge for future
research is to study several ecosystems processes
concurrently and to investigate how these processes
are interdependent.
Potential drivers of SR
Within habitat types, SR exhibited distinct diel and
seasonal patterns and was strongly correlated with
temperature. Temperature has been identified as one
of the most important predictors for SR (Gansert,
1994; Burton et al., 1998; Buchmann, 2000). The average Q10 value in the Tagliamento floodplain was
similar to other terrestrial ecosystems (Davidson et al.,
1998; Buchmann, 2000; Raubuch et al., 2002; Chivers
et al., 2009), although differences among habitat types
were distinct (1.62–4.57; Tables 4 & 5). SR represented
the sum of autotrophic (root) and heterotrophic
respiration, for which the Q10 can be quite different
(Buchmann et al., 1997; Boone et al., 1998). In most
habitats, the temperature-corrected (standardised at
12 C) seasonal trend in SR reflected the potential
influence of root respiration, with higher respiration
during the growing season (Fig. 3; Buchmann, 2000;
Hanson et al., 2000). In the riparian forest, however,
the standardised respiration peaked in January, and
not in July, like in the other floodplain habitats
investigated. To our knowledge, the lack of seasonality in forest soil respiration is not supported by other
studies. We cannot rule out that the temperature
dependence function (eqn 3) may fail to estimate the
potential respiration of a soil community at 12 C if
the calculation is based on respiration measured
around freezing point or that differences in organic
matter quality may account for the higher respiration.
However, we lack information to support or reject
these assumptions.
Frequent flood disturbance (Jones, Fisher &
Grimm, 1995; Uehlinger, 2000, 2006) and high-temperature amplitudes that can affect abundance,
physiology and composition of microbial communities (Lloyd & Taylor, 1994; Palmer Winkler, Cherry &
Schlesinger, 1996; Rinklebe & Langer, 2006) might
explain the minor temperature dependence of SR in
channel habitats and exposed gravel sediments
(Table 5). Periods of heterotrophic activity do not
necessarily coincide with periods of autotrophic
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
Sediment respiration in a floodplain mosaic
activity (Acuña et al., 2004). It is the interplay
between these processes that determine the seasonality of SR, and hence, differences among habitat
types.
In addition to temperature, moisture (Davidson
et al., 2000; Xu & Qi, 2001; Li et al., 2006; McIntyre
et al., 2009) and OM content (Schimel et al., 1994;
Randerson et al., 1996) are known to influence SR. For
example, Carlyle & BaThan (1988) found a significant
correlation between SR and moisture when water
content dropped below 12.5%. Amalfitano et al. (2008)
found that a water content of 20% (% of max. WHC;
grain size <2 mm) was the critical threshold for
bacterial carbon production (BCP). Below this value,
BCP ceased, although a considerable proportion of
bacteria remained alive. In our floodplain system,
water content was always well above this critical
threshold level (Table 2). Although OM and water
content did not explain within-habitat variation of SR,
OM content was closely linked to differences in SR
among habitat types. Factors such as OM quality,
water availability and root density, which were not
assessed in the present study but are known to be
highly variable in space (Buchmann, 2000), may
contribute to the observed within-habitat variability
of SR.
Flooding influences soil and sediment respiration.
However, the impact of flooding could not be
assessed because there were no major floods during
our investigation and small flood pulses did not affect
the terrestrial sampling sites. Most rewetting studies
in permanent terrestrial habitats (e.g. Orchard &
Cook, 1983; Carlyle & BaThan, 1988; Fierer & Schimel,
2002; Fischer, 2009) resulted in increased SR, whereas
flooding (WC > WHC) often causes a decrease in SR
(Guntiñas et al., 2009; McIntyre et al., 2009). In contrast, Valett et al. (2005) found a >200-fold increase in
soil respiration after a few days of inundation in an
experimentally flooded floodplain forest of the Middle Rio Grande. In the Tagliamento floodplain, flow
pulses that inundated up to 20% of the active tract
lasted on average <6 days, and floods that affected
80% of the active tract lasted on average <1.5 days
(Van der Nat et al., 2002). Furthermore, at base flow
conditions, most of the floodplain consists of exposed
gravel sediments ( 53%) and permanent aquatic
habitats (16%), which exhibited lowest SR. Therefore, the effect of flooding on annual SR is probably
low.
11
Annual soil and sediment CO2 flux
Total mean annual SR (area-weighted mean) for the
entire island-braided floodplain was calculated as
467 g C m)2 year)1 and is similar to the mean annual
respiration reported for temperate wetlands (Table 7).
However, this mean value does not take into consideration the high spatial complexity in SR within and
among the various habitat types in the Tagliamento
floodplain. Habitat-specific efflux of CO2 from soils
and sediments spanned almost the entire range
reported from aquatic and terrestrial ecosystems
globally. While exposed gravel sediments and pond
habitats exhibited CO2 fluxes similar to Arctic and
desert ecosystems, fluxes from vegetated islands were
as high as in tropical forests (Table 7). Hence, in the
natural floodplain mosaic of the Tagliamento River,
‘desert-like’ habitats are located in close proximity to
‘rainforest-like’ habitats. The abrupt change in respiratory carbon fluxes at the interface between vegetated (islands and riparian forest) and unvegetated
habitat types (exposed gravel sediments and aquatic
habitats) indicates not only fundamental differences
in C sources and C cycling within these contrasting
habitat types but also limited transfer of C across
habitat boundaries. The direct input of litter to the
active tract is restricted to the immediate interface
between vegetated islands, the riparian forest and the
active tract (Langhans, 2006).
Methodological aspects in assessing SR
Unfortunately, it was not possible to apply the same
methods for quantifying SR in aquatic and terrestrial
habitat types. This difference constrained to some
extent comparability between aquatic and terrestrial
habitat types. The assessment of terrestrial respiration
provided area-specific values, and sediment or soil
patches remained relatively undisturbed during sampling. In aquatic systems, enclosures are a common
method used to quantify SR (Pusch & Schwoerbel,
1994; Jones, 1995; Jones et al., 1995; Naegeli et al.,
1995). Enclosures create an artificial sediment structure that may disrupt the enclosed microbial communities (Uehlinger & Brock, 1991; Naegeli et al., 1995).
Moreover, the conversion of SR from chamber to
area-related values required information about the
change in SR with sediment depth, a difficult task in
aquatic environments (Naegeli & Uehlinger, 1997).
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
12
M. Doering et al.
Open-system methods (Odum, 1956) can circumvent
such methodological constraints. In braided systems,
however, the divergence and convergence of channels
around bars and islands and the intense vertical
exchange of surface and ground waters prevented
determination of reliable oxygen balances. By integrating the measured aquatic respiration over a depth
of 20 cm, we probably underestimated aquatic sediment respiration.
In terrestrial habitats, SR measurements using
IRGA probably integrated the SR in the uppermost
20–40 cm of soil and sediment layers (Luo & Zhou,
2006). Furthermore, respiration included different
functional compartments. In aquatic habitats, respiration included only the contribution of the hyporheic
zone. In gravel deposits and large accumulations of
wood, where an autotrophic compartment was almost
completely missing, SR presumably represented
microbial respiration. However, in islands and riparian forest, SR encompassed both microbial and autotrophic, i.e. root respiration. Root ⁄ rhizosphere
respiration can account for as little as 10% to more
than 90% of total in situ soil respiration depending on
vegetation type and season and is usually highest
during the growing seasons (Hanson et al., 2000).
Further, heterotrophic abundance, composition and
activity of fungi and bacteria is habitat specific (e.g.
type of soil and sediment) and depends on seasonal
variations in habitat properties (Battin et al., 2004;
Rinklebe & Langer, 2006). Although respective seasonal investigations into floodplain soils and sediments are lacking, it can be assumed that seasonal
contribution of heterotrophic organisms to SR varies
differently in individual habitat types throughout the
year. Despite these methodological shortcomings, our
study provides a first estimate of the spatiotemporal
heterogeneity of SR in a highly complex and dynamic
river landscape.
In conclusion, we found a tight link between
habitat properties and soil and sediment respiration
across the aquatic–terrestrial environment. SR differed by an order-of-magnitude among habitat types,
with temperature being the most important variable
determining SR. Hence, even small changes in the
relative proportion of individual habitat types and
temperature can substantially alter whole-floodplain
efflux of CO2. This is important in the context of
river modification and changing climate. Channelisation and flow regulation decrease habitat complexity
in floodplain systems, with severe consequences for
the spatial variation in SR. For example, vegetated
islands are among the first landscape elements that
disappear as a consequence of flow regulation and
channel modification (Gurnell et al., 2001). The
clearance of all vegetated islands (9% of total area;
23% of total floodplain SR), as was the case during
World Wars I and II, would lead to a massive
decrease in the overall floodplain SR. On the other
hand, a lack of hydrological dynamics could lead to
an extension of the relative proportion of the
floodplain forest (23% of total area; 48% of total
floodplain SR) with a subsequent increase in SR
(Table 6). Global warming will increase aquatic and
terrestrial soil and sediment temperature and therefore SR. As a consequence, decomposition and
mineralisation can exceed net primary production
and deposition leading to a decrease in soil and
sediment organic matter content (Kirschbaum, 1995;
Gudasz et al., 2010), and therefore, decrease process
heterogeneity and probably carbon cycling in floodplain systems.
Acknowledgments
The authors thank the many people who helped in the
field and with logistics, especially C. Cruciat, S. Blaser,
Y. Schwill, S. Ismail, M. Alp, S. Frey and E. van
Daalen. Thanks to L. Indermaur for providing GIS
data. The work was supported by the EU-project
tempQsim (EVK1-CT2002-00112; http://www.tempqsim.net) and by SBF (No. 02.0072). We are grateful to
previous reviewers for their comments that considerably improved the manuscript.
References
Acuña V., Giorgi A., Muños I., Uelinger U. & Sabater S.
(2004) Flow extremes and benthic organic matter shape
the metabolism of a headwater Mediterranean stream.
Freshwater Biology, 49, 960–971.
Amalfitano S., Fazi S., Zoppini A., Caracciolo A.B.,
Grenni P. & Puddu A. (2008) Responses of benthic
bacteria to experimental drying in sediments from
Mediterranean temporary rivers. Microbial Ecology, 55,
270–279.
Andrews J.A. & Schlesinger W.H. (2001) Soil CO2
dynamics, acidification, and chemical weathering in a
temperate forest with experimental CO2 enrichment.
Global Biogeochemical Cycles, 15, 149–162.
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
Sediment respiration in a floodplain mosaic
Arscott D.B., Tockner K., Van der Nat D. & Ward J.V.
(2002) Aquatic habitat dynamics along a braided
alpine river ecosystem (Tagliamento River, Northeast
Italy). Ecosystems, 5, 802–814.
Ball B.A., Virginia R.A., Barrett J.E., Parsons A.N. & Wall
D.H. (2009) Interactions between physical and biotic
factors influence CO2 flux in Antarctic dry valley soils.
Soil Biology and Biochemistry, 41, 1510–1517.
Battin T.J., Wille A., Psenner R. & Richter A. (2004) Largescale environmental controls on microbial biofilms in
high-alpine streams. Biogeosciences, 1, 159–171.
Bertoldi W., Gurnell A.M., Surian N., Tockner K., Zanoni
L., Ziliani L. et al. (2009) Linking hydrology, geomorphology and ecology of a multi-thread river: time
scales of variability and change on the Tagliamento
River as a reference for river restoration. River Research
and Applications, 25, 501–516.
Bond-Lamberty B.P. & Thomson A.M. (2010) A global
database of soil respiration data, Version 1.0. Data set.
Available on-line [http://daac.ornl.gov] from Oak
Ridge National Laboratory Distributed Active Archive
Center, Oak Ridge, Tennessee, U.S.A. doi:10.3334 ⁄
ORNLDAAC ⁄ 984.
Boone R.D., Nadelhofer K.J., Canary J.D. & Kaye J.P.
(1998) Roots exert a strong influence on the temperature sensitivity of soil respiration. Nature, 396, 570–572.
Buchmann N. (2000) Biotic and abiotic factors controlling
soil respiration rates in Picea abies stands. Soil Biology
and Biochemistry, 32, 1625–1635.
Buchmann N., Guehl J.M., Barigah T.S. & Ehleringer J.R.
(1997) Interseasonal comparison of CO2 concentrations, isotopic composition, and carbon dynamics in an
Amazonian rainforest (French Guiana). Oecologia, 110,
120–131.
Burton A.J., Pregitzer K.S., Zogg G.P. & Zak D.R. (1998)
Drought reduces root respiration in sugar maple
forests. Ecological Applications, 8, 771–778.
Carlyle J.C. & BaThan U. (1988) Abiotic controls of soil
respiration beneath an eighteen year old Pinus radiata
stand in South-eastern Australia. Journal of Ecology, 76,
654–662.
Chivers M.R., Turetsky M.R., Waddington J.M., Harden
J.W. & McGuire A.D. (2009) Effects of experimental
water table and temperature manipulations on ecosystem CO2 fluxes in an Alaskan rich fen. Ecosystems, 12,
1329–1342.
Cuffney T.F. (1988) Input, movement and exchange of
organic matter within a subtropical coastal blackwater
river-floodplain system. Freshwater Biology, 19, 305–
320.
Davidson E.A., Belk E. & Boone R.D. (1998) Soil
water content and temperature as independent or
confounded factors controlling soil respiration in a
13
temperate mixed hardwood forest. Global Change Biology, 4, 217–227.
Davidson E.A., Verchot L.V., Cattânio J.H., Ackerman
I.L. & Carvalho J.E.M. (2000) Effects of soil water
content on soil respiration in forests and cattle pastures
of eastern Amazonia. Biogeochemistry, 48, 53–69.
Dilly O. (2001) Microbial respiratory quotient during
basal metabolism and after glucose amendment in
soils and litter. Soil Biology and Biochemistry, 33, 117–
127.
Eglin S.W.T. (1990). Die Zusammensetzung und kleinräumige Verteilung der Makroinvertebratenzoenose eines natürlichen, voralpinen Fliessgewässers (Thur) in Abhängigkeit
vom Nahrungsangebot und der Sedimentstruktur. PhD
Thesis. ETH, Zürich.
Euskirchen E.S., Jiquan C., Gustafson E.J. & Siyan M.
(2003) Soil respiration at dominat patch types within a
managed Northern Wisconsin landscape. Ecosystems, 6,
595–607.
Fierer N. & Schimel J.P. (2002) Effects of drying-rewetting frequency on soil carbon and nitrogen transformations. Soil Biology and Biochemistry, 34, 777–787.
Fischer T. (2009) Substantial rewetting phenomena on
soil respiration can be observed at low water availability. Soil Biology and Biochemistry, 41, 1577–1579.
Gansert D. (1994) Root respiration and its importance for
the carbon balance of beech seedlings (Fagus sylvatica
L.) in a montane beech forest. Plant and Soil, 167, 109–
119.
del Giorgio P.A. & Williams P.J.B. (2005) Respiration in
Aquatic Ecosystems. Oxford ⁄ University Press, Oxford.
Gudasz C., Bastviken D., Steger K., Premke K., Sobek S.
& Tranvik L.J. (2010) Temperature-controlled organic
carbon mineralization in lake sediments. Nature, 466,
478–481.
Guntiñas M.E., Gil-Sotres F., Leirós M.C. & TrasarCepeda C. (2009) CO2 emission from soils under
different uses and flooding conditions. Soil Biology
and Biochemistry, 41, 2598–2601.
Gurnell A.M., Petts G.E., Hannah D.M., Smith B.P.G.,
Edwards P.J., Kollmann J. et al. (2001) Riparian vegetation and island formation along the gravel-bed
Fiume Tagliamento, Italy. Earth Surface Processes and
Landforms, 26, 31–62.
Hanson P.J., Edwards N.T., Garten C.T. & Andrews J.A.
(2000) Separating root and soil microbial contributions
to soil respiration: a review of methods and observations. Biogeochemistry, 48, 115–146.
Ingendahl D., Borchardt D., Saenger N. & Reichert P.
(2009) Vertical hydraulic exchange and the contribution of hyporheic community respiration to whole
ecosystem respiration in the River Lahn (Germany).
Aquatic Science, 71, 399–410.
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
14
M. Doering et al.
Janssens I.A., Lankreijer H., Matteucci G., Kowalski A.S.,
Buchmann N., Epron D. et al. (2001) Productivity
overshadows temperature in determining soil and
ecosystem respiration across European forests. Global
Change Biology, 7, 269–278.
Jones J.B. (1995) Factors controlling hyporheic respiration
in a desert stream. Freshwater Biology, 34, 91–99.
Jones J.B., Fisher S.G. & Grimm N.B. (1995) Vertical
hydrologic exchange and ecosystem metabolism in a
Sonoran desert stream. Ecology, 76, 942–952.
Junk W.J., Bayley P.B. & Sparks R.E. (1989) The flood
pulse concept in river-floodplain systems. Canadian
Special Publications for Fisheries and Aquatic Science, 106,
110–126.
Jussel P. (1992) Modellierung des Transports gelöster Stoffe
in inhomogenen Grundwasserleitern. PhD Thesis. ETH,
Zürich.
Karrenberg S., Kollmann J., Edwards P.J., Gurnell A.M. &
Petts G.E. (2003) Patterns in woody vegetation along
the active zone of a near-natural Alpine river. Basic and
Applied Ecology, 4, 157–166.
Kirschbaum M.U.F. (1995) The temperature dependence
of soil organic matter decomposition and the effect of
global warming on soil organic carbon storage. Soil
Biology and Biochemistry, 27, 753–760.
Langhans S.D., (2006) Riverine floodplain heterogeneity
as a controller of the organic matter dynamics and
terrestrial invertebrate distribution. PhD Thesis. ETH,
Zürich.
Langhans S.D., Tiegs S.D., Uehlinger U. & Tockner K.
(2006) Environmental heterogeneity controls organicmatter dynamics in river floodplain ecosystems. Polish
Journal of Ecology, 54, 675–680.
Langhans S.D., Tiegs S.D., Gessner M.O. & Tockner K.
(2008) Leaf-decomposition heterogeneity across a riverine floodplain mosaic. Aquatic Sciences, 70, 337–346.
Li Y.Q., Xu M. & Zou X.M. (2006) Heterotrophic soil
respiration in relation to environmental factors and
microbial biomass in two wet tropical forests. Plant and
Soil, 281, 193–201.
Lloyd J. & Taylor J.A. (1994) On the temperature
dependence of soil respiration. Functional Ecology, 8,
315–323.
Luo Y. & Zhou X. (2006) Soil Respiration and the Environment. Academic Press, Amsterdam.
McIntyre R.E.S., Adams M.A., Douglas J.F. & Grierson
P.F. (2009) Rewetting and litter addition influence
mineralization and microbial communities in soils
from a semi-arid intermittent stream. Soil Biology and
Biochemistry, 41, 92–101.
Naegeli M.W. & Uehlinger U. (1997) Contribution of
the hyporheic zone to ecosystem metabolism in a
prealpine gravel-bed river. Journal of the North American Benthological Society, 16, 794–804.
Naegeli M.W., Hartmann U., Meyer E.I. & Uehlinger U.
(1995) POM-dynamics and community respiration in
the sediments of a floodprone prealpine river (Necker,
Switzerland). Archiv für Hydrobiologie, 133, 339–347.
Naiman R.J., Décamps H. & McClain M.E. (2005) Riparia:
Ecology, Conservation and Management of Streamside
Communities. Elsevier ⁄ Academic Press, San Diego.
Norman J.M., Kucharik C.J., Gower S.T., Baldocchi D.D.,
Crill P.M., Rayment M. et al. (1997) A comparison of
six methods for measuring soil-surface carbon dioxide
fluxes. Journal of Geophysical Research-Atmospheres, 102,
28771–28777.
Odum H.T. (1956) Primary production in flowing waters.
Limnology and Oceanography, 1, 102–117.
Orchard V.A. & Cook F.J. (1983) Relationship between
soil respiration and soil moisture. Soil Biology and
Biochemistry, 15, 447–453.
Palmer Winkler J., Cherry R.S. & Schlesinger W.H. (1996)
The Q10 relationship of microbial respiration in a
temperate forest soil. Soil Biology and Biochemistry, 28,
1067–1072.
Petts G.E., Gurnell A.M., Gerrard A.J., Hannah D.M.,
Hansford B., Morrissey I. et al. (2000) Longitudinal
variations in exposed riverine sediments: a context for
the ecology of the Fiume Tagliamento, Italy. Aquatic
Conservation: Marine and Freshwater Ecosystems, 10, 249–
266.
Pusch M. & Schwoerbel J. (1994) Community respiration
in hyporheic sediments of a mountain stream (Steina,
Black Forest). Archiv für Hydrobiologie, 130, 35–52.
Raich J.W. & Schlesinger W.H. (1992) The global carbon
dioxide flux in soil respiration and its relationship to
vegetation and climate. Tellus, 44B, 81–99.
Randerson J.T., Matthew V.T., Malmstrom C.M., Field
B.F. & Fung I.Y. (1996) Substrate limitations for
heterotrophs: implications for models that estimate
the seasonal cycle of atmospheric CO2. Global Biogeochemical Cycles, 10, 585–602.
Raubuch M., Dyckmans J., Joergensen R.G. & Kreutzfeldt
M. (2002) Relation between respiration, ATP content,
and Adenylate Energy Charge (AEC) after incubation
at different temperatures after drying and rewetting.
Journal of Plant Nutrition and Soil Science, 165, 435–440.
Rinklebe J. & Langer U. (2006) Microbial diversity in
three floodplain soils at the Elbe River (Germany). Soil
Biology and Biochemistry, 38, 2144–2151.
Schimel D.S., Braswell B.H., Holland E.H., McKeown R.,
Ojima D.S., Painter T.H. et al. (1994) Climatic, edaphic,
and biotic controls over storage and turnover of carbon
in soils. Global Biogeochemical Cycles, 8, 279–293.
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x
Sediment respiration in a floodplain mosaic
Sokal R.R. & Rohlf F.J. (1995) Biometry: The Principles and
Practice of Statistics in Biological Research, 3rd edn.
Freeman, New York.
Tang X.L., Zhou G.Y., Liu S.G., Zhang D.Q., Liu S.Z., Li J.
et al. (2006) Dependence of soil respiration on soil
temperature and soil moisture in successional forests
in southern China. Journal of Integrative Biology, 48, 654–
663.
Tockner K. & Stanford J.A. (2002) Riverine flood plains:
present state and future trends. Environmental Conservation, 29, 308–330.
Tockner K., Malard F. & Ward J.V. (2000) An extension of
the flood pulse concept. Hydrological Processes, 14,
2861–2883.
Tockner K., Ward J.V., Arscott D.B., Edwards P.J.,
Kollmann J., Gurnell A.M. et al. (2003) The Tagliamento River: a model ecosystem of European importance.
Aquatic Science, 65, 239–253.
Tockner K., Klaus I., Baumgartner C. & Ward J.V. (2006)
Amphibian diversity and nestedness in a dynamic
floodplain river (Tagliamento, NE-Italy). Hydrobiologia,
565, 121–133.
Tockner K., Lorang M.S. & Stanford J.A. (2010) River
floodplains as model ecosystems to test general
hydrogeomorphic and ecological concepts. River Research and Applications, 26, 76–86.
Tonolla D., Acuna V., Uehlinger U., Frank T. & Tockner
K. (2010) Thermal heterogeneity in river floodplains.
Ecosystems, 13, 727–740.
Uehlinger U. (1993) Primary production and respiration
in the outlet of an eutrophic lake (River Glatt,
Switzerland). Archiv für Hydrobiologie 128, 39–55.
Uehlinger U. (2000) Resistance and resilience of ecosystem metabolism in a flood-prone river system. Freshwater Biology, 45, 319–332.
Uehlinger U. (2006) Annual cycle and inter-annual
variability of gross primary production and ecosystem
respiration in a floodprone river during a 15-year
period. Freshwater Biology, 51, 938–950.
15
Uehlinger U. & Brock J.T. (1991) The assessment of river
periphyton metabolism: a method and some problems.
In: Use of Algae for Monitoring Rivers (Eds B.A. Whitton,
E. Rott & G. Friedrich), pp. 175–181. University
Innsbruck, Institute für Botanik, Innsbruck, Austria.
Uehlinger U. & Naegli M.W. (1998) Ecosystem metabolism, disturbance, and stability in a prealpine gravel
bed river. Journal of the North American Benthological
Society, 17, 165–178.
Uehlinger U., Naegeli M.W. & Fisher S.G. (2002) A
heterotrophic desert stream? the role of sediment
stability. Western North American Naturalist, 62, 466–473.
Valett H.M., Baker M.A., Morrice J.A., Crawford C.S.,
Molles M.C., Dahm C.N. et al. (2005) Biogeochemical
and metabolic responses to the flood pulse in a
semiarid floodplain. Ecology, 86, 220–234.
Van der Nat D., Schmidt A.P., Tockner K., Edwards P.J.
& Ward J.V. (2002) Inundation dynamics in braided
floodplains: Tagliamento River, northeast Italy. Ecosystems, 5, 636–647.
Van der Nat D., Tockner K., Edwards P.J. & Ward J.V.
(2003) Habitat change in braided floodplains (Tagliamento, NE-Italy). Freshwater Biology, 48, 1799–1812.
Van Voris P., O’Neill R.V., Emanuel W.R. & Shugart H.H.
(1980) Functional complexity and ecosystem stability.
Ecology, 61, 1352–1360.
Ward J.V., Tockner K., Edwards P.J., Kollmann J.,
Bretschko G., Gurnell A.M. et al. (1999) A reference
system for the Alps: the ‘‘Fiume Tagliamento’’. Regulated Rivers: Research & Management, 15, 63–75.
Ward J.V., Tockner K., Arscott D.B. & Claret C. (2002)
Riverine landscape diversity. Freshwater Biology, 47,
517–539.
Xu M. & Qi Y. (2001) Soil-surface CO2 efflux and its
spatial and temporal variations in a young ponderosa
pine plantation in northern California. Global Change
Biology, 7, 667–677.
(Manuscript accepted Date: 20 December 2010)
2011 Blackwell Publishing Ltd, Freshwater Biology, doi:10.1111/j.1365-2427.2011.02569.x